This invention relates to a method and apparatus for identifying the sampling phase with respect to the colour sub-carrier of a digital television system and is particularly, but not solely applicable to a digital time base corrector of the kind disclosed in co-pending U.S. patent applications Ser. Nos. 558,091 and 558,090.
In a digital television system using the NTSC colour system, it is often desirable to have the sampling frequency equal to three times the colour sub-carrier frequency. However, it is also often desirable that the sampling be coherent on a line to line basis, i.e. the sample points align vertically in the picture. In order to achieve this with the NTSC colour system, the phase of the clock frequency locked to three times sub-carrier frequency is alternated by 180.degree. line to line to produce the required line coherent sampling pulses. If the sampled digital data is then written into a digital storage system and subsequently read from that storage system under the control of similar line coherent clock pulses, then it is desirable under certain circumstances that the two possible phase relationships of the Write and Read clock pulses to the colour sub-carrier frequency are identified on the input and output side of the storage and can be matched together.
One circumstance under which it is important to identify the sampling phase is when the colour sub-carrier burst is not written into the storage along with the active line time. When this is the case, the burst is often re-inserted on the Read side of the store in some predetermined phase relationship with the output signal. To make sure the burst to chrominance phase relationship is not changed by 180.degree. after a disturbance to the input signal or to the locking signal controlling the Read operation, it is necessary to perform the afore-mentioned phase identification and input/output matching.